• Laser & Optoelectronics Progress
  • Vol. 55, Issue 6, 061011 (2018)
Shanxin Zhang, Qiang Fan*; , and Zhiping Zhou
Author Affiliations
  • Engineering Research Center of Internet of Things Technology Applications, Ministry of Education, Jiangnan University, Wuxi, Jiangsu 214000, China
  • show less
    DOI: 10.3788/LOP55.061011 Cite this Article Set citation alerts
    Shanxin Zhang, Qiang Fan, Zhiping Zhou. Object Shape Classification Based on Bayesian Optimized Neural Network[J]. Laser & Optoelectronics Progress, 2018, 55(6): 061011 Copy Citation Text show less
    Algorithmic framework
    Fig. 1. Algorithmic framework
    Examples of Animal dataset
    Fig. 2. Examples of Animal dataset
    80 objects in ETH-80 dataset
    Fig. 3. 80 objects in ETH-80 dataset
    Comparison of proposed method and pSGLD method
    Fig. 4. Comparison of proposed method and pSGLD method
    MethodAccuracyPrecisionRecallF1
    BCF[6]83.2083.1982.8583.02
    Bioinformaticsapproach[11]83.7083.6783.6083.63
    pSGLD[14]83.9083.9884.0083.83
    Proposed84.6084.6784.6084.63
    Table 1. Results of the Animal dataset%
    MethodBirdButterflyCatCowCrocodileDeerDogDolphinDuckElephant
    BCF[6]87.692.273.877.474.890.482.689.087.095.2
    pSGLD[13]87.897.476.387.272.889.784.478.288.095.5
    Proposed88.097.876.087.872.589.786.280.088.395.5
    MethodFishFly-birdHenHorseLeopardMonkeyRabbitRatSpiderTortoise
    BCF[6]79.870.094.295.466.458.485.870.699.293.6
    pSGLD[13]81.868.291.496.067.970.291.164.510090.0
    Proposed81.868.291.595.870.071.992.168.210090.5
    Table 2. Detailed accuracy of each method in Animal Dataset%
    MethodAccuracyPrecisionRecallF1
    BCF[6]90.2290.0389.9289.97
    Bioinformaticsapproach[11]90.6890.4790.3490.40
    pSGLD[14]91.6691.6291.6991.65
    Proposed92.0592.0892.1092.09
    Table 3. Classification results in ETH-80 dataset%
    Shanxin Zhang, Qiang Fan, Zhiping Zhou. Object Shape Classification Based on Bayesian Optimized Neural Network[J]. Laser & Optoelectronics Progress, 2018, 55(6): 061011
    Download Citation